21 Transmission Line Distance Prot II r8
21 Transmission Line Distance Prot II r8
6022.pdf, Z = V/I Does Not Make a Distance Relay by J. Roberts; A. Guzman; E.O.
Schweitzer, III
6010.pdf, Distance Relay Element Design by Jeff Roberts; Dr. Edmund O. Schweitzer
III
6065.pdf, Application Guidelines for Ground Fault Protection by Joe Mooney, P.E.;
Jackie Peer
l Infeed
l Fault Resistance
l Unequal Measured Impedances
During Faults
l Evolving Faults
l Load Encroachment
l Mutual Coupling
l Simultaneous Faults
l Cross-Country Faults
l Power Swings
l Three-Terminal Lines
All these problems may affect distance and directional overcurrent relays. Cross-country
faults, simultaneous faults, and CT saturation may also present a problem for differential
schemes.
l Reactor-Compensated Lines
l Short Lines
l CT Saturation
l CCVT Transients
l Series-Compensated Lines
Three-terminal lines and short lines also have special protection requirements. The
ringdown at subharmonic frequency resulting from compensation reactors may also
create protection problems. Series-compensated lines are extremely difficult to protect.
All protection principles may have problems, because of the possibility of voltage and
current inversions. If the series-compensation capacitors are carefully selected, the
possibility of current inversions can be eliminated. In such a case, a differential
protection scheme may be the best option.
In this presentation, we will examine transmission line protection problems that exclude
CT saturation, CCVT transients, series-compensated lines, three-terminal lines, short
lines, and reactor-compensated lines.
Fault Resistance
l Phase Faults
t Arc resistance
t Other components (trees, etc.)
l Ground Faults
t Arc resistance
t Tower and tower footing
t Ground return path
t Other components (trees, etc.)
Fault resistance affects all protection principles to some extent. For phase faults (three-
phase, line-line) fault resistance results largely from the resistance of the arc between the
faulted conductors. If the fault is initiated by a tree or something else in the line, its
resistance should also be considered.
Ground fault resistance includes the resistance of the arc between the conductor and the
tower, the tower and tower footing resistance, and the ground return path resistance.
Ground faults may also involve other objects such as trees.
Ground fault resistance values are typically much greater than phase fault resistances. For
transmission line faults involving trees, for example, the fault resistance may be on the
order of hundreds of ohms.
Arc Resistance
440 × L
R arc = W
I
L: Arc Length in Feet
I: Arc Current in Amperes
Arc resistance is quite variable. A commonly accepted value for currents between 70 A
and 20,000 A is an arc voltage drop of 440 V per phase, independent of current
magnitude.
Arc Resistance
28710 × L
Rarc = 1.4
Ω
I
L: Arc Length in Meters
I: Arc Current in Amperes
This is another empirical expression for arc resistance with the arc length in meters
(instead of feet). Observe that there is a 1.4 exponent in the current in this expression.
Fault Resistance
Radial Lines
V mZL
I R
F
21
Z = V I = mZ L + R F
The figure shows the effect of fault resistance on the impedance a distance element
measures in a radial system. The distance element measures the fault loop impedance,
including the fault resistance. The result is a distance estimate greater than the real
distance to the fault. This inaccurate distance estimate makes the distance element
underreach.
Fault Resistance
Looped Lines
V mZL
I RF
IF
21
V = I mZL + IFRF
IF
Z = V I = mZ L + RF
I
An additional problem in looped lines is the infeed effect in the fault impedance. The
relay does not measure the current contribution to the fault from the remote-end source.
As a result, RF in the impedance estimate is multiplied by an infeed effect factor IF / I.
The effect of this factor is twofold. The infeed effect increases the value of the apparent
fault resistance (the magnitude of the factor is greater than unity). The infeed effect factor
is, in general, a complex number, so the apparent fault impedance is no longer purely
resistive.
X
IF
RF
I
mZL
Z
In this impedance plane representation, the effect of fault resistance in looped lines can
be seen. The distance element should measure an impedance mZL. However, the
measured impedance is Z. Observe that the infeed effect factor, IF / I, increases the value
and produces a phase shift in the fault resistance RF.
E Ðd BUS S I BUS R E Ð0
ZS m•ZL (1-m)•ZL ZR
RELAY
RF IF
The power system model shown will be used to study the effect of RF and the power
angle, d, on the apparent fault impedance. For simplicity, a homogeneous system (all
source and line impedances have the same angle) will be considered.
Effect of RF and δ
Im (V/I)
10
RF (W sec.)
8 d = -60°
Reactance (W sec.)
X =0
6 d = -30° =1
=4
4 X d = 0°
=8
2 d = +30°
Z1L
d = +60°
0 Re (V/I)
-2
-2 0 2 4 6 8 10 12 14
Resistance (W sec.)
This figure shows the effect of RF and d on the impedance estimate. For bolted faults,
the distance element measures the correct impedance value. An increase in the value
of RF increases the measured impedance and produces relay underreach.
mZL
V
I IF
21
ZX C = I IF
A distance protection scheme has six basic relay elements. For the phase elements, the
line-line voltages and the differences of the line currents are used as input signals.
Ground distance elements receive the phase voltages and the compensated line currents
as input signals. The zero-sequence current is used to compensate the line current inputs
of ground distance elements. These connections ensure that the fault-loop element(s)
correctly measure the fault-loop impedance. For example, for an ABG fault, three
distance elements correctly estimate impedance: AB, AG, and BG elements. The
question is, what impedances do the other three elements measure for this fault? These
elements need to measure impedances with values no lower than the fault-loop
impedance. This ensures that the distance elements that measure the correct impedance
value will make the tripping decision.
The simple power system shown in the figure can be used for an analytical study of the
impedances measured by the different distance elements during faults. The idea is to
derive the expressions of the measured impedance using symmetrical component
techniques. In the figure, mZL is the impedance of the protected line section. The
positive-sequence value of this impedance, mZL1, is the correct value that the distance
elements measure. ZX is an impedance including mZL and the source impedance behind
the relay. The factor C expresses the infeed effect in the fault resistance.
Measured Impedances
This table shows the expressions of the impedances phase distance elements measure for
two types of phase faults (ABC and BC faults). For ABC faults, all three elements
measure the loop impedance value, which unfortunately includes the fault impedance
term. For BC faults, only the fault-loop element BC correctly measures the impedance.
The other two elements (AB and CA) estimate higher impedances. This means that the
BC element will make the tripping decision for zone-end faults.
Measured Impedances
RF Z X1 R
CG m Z L1 + m Z L1 + Ð90° + F Ð30°
C1 3 3C1
This table shows the expressions of the impedances measured by ground distance
elements for two types of phase faults.
For ABC faults, all three elements measure the fault-loop impedance. In other words,
ground distance elements may respond to three-phase faults unless, for example, their
operation is supervised with a zero-sequence overcurrent element. This tendency to
operate is also present for other system balanced conditions such as normal load or power
swings. For phase-to-phase faults, the ground distance elements measure impedance
values greater than the fault-loop impedance.
The effect of ground faults on ground distance elements is not presented in this analysis.
The result is the same as for the phase elements. That is, the fault-loop element correctly
measures the impedance, and the healthy phase elements estimate impedance values
greater than the fault-loop impedance.
m (per-unit of ZL)
ZS ZL
SOURCE S Relay
FAULT
ZS1 = 1 Ohm ZL1 = 8 Ohms
ZS0 = 3 Ohms ZL0 = 24 Ohms
Digital simulation is an excellent tool for studying the impedances that distance elements
measure during faults in complex power systems. The figure shows a power system
model that will be used for digital simulation studies.
Characteristics
Self-Polarized
Expanded
4
4 CG 2 AB
0
0 AG
R
0 0 CA
4
BG
4
The figure shows the impedances all distance elements measure for a close-in A-phase-
to-ground fault for different RF values (0 to 4 ohms).
The figure indicates that only the AG distance element correctly measures the impedance
for this AG fault. It is also clear from the figure that a self-polarized mho element may
not see the fault. For a polarized mho element, the fault is well within the characteristic.
Other distance elements such as AB, CG, and CA may operate for this fault, so distance
element operation cannot be relied on to make single-pole tripping decisions. In this case,
for example, a three-pole trip would be issued instead of a single-pole trip for a single-
line-to-ground fault.
100%
50%
CA
AG
50%
CG 0 0
R
0 0
50% 50%
100%
100%
BG
The figure shows the effect of displacing the fault location along the protected line. The
result is the appearance of fault areas in the impedance plane. The impedance each
distance element measures will lie inside the corresponding fault area. The position for
the measurement depends on fault location and fault resistance. The sides of the fault
areas marked with dots are the locii of the measured impedances for bolted (RF = 0)
faults. The straight lines parallel to those sides are the locii of the measured impedances
corresponding to the maximum RF value (4 ohms.)
The basic conclusion is the same as for the previous figure: there are several distance
elements prone to operate for a single line-to-ground fault. For single-pole tripping, the
tripping decisions of the distance elements cannot be used. A separate algorithm is
needed to determine the fault type for single-pole applications.
This figure shows the impedances the distance elements measure for a BCG fault at the
remote end of the relay reach. The dots on each fault locus represent different values of
fault resistance (0 to 10 ohms).
For bolted faults, three distance elements (BC, BG, and CG) see the fault exactly at the
end of the reach corresponding to each element. When the fault resistance increases, the
impedance the CG element measures moves away from the relay characteristic. On the
other hand, the BC impedance penetrates the relay characteristic. This element
overreaches, so it must be blocked from operation for this fault. For simplicity, the effect
of the fault resistance on the BC element impedance is not shown. It is clear, however,
that this impedance also leaves the operation characteristic.
Evolving Faults
l Variable RF and IF
l The Fault Type Changes
(AG®ABG ®ABCG)
l It Is Difficult to Detect Fault
Inception and Fault Type
Changes
Evolving faults present problems for all protection principles. Many faults evolve in
some way. The fault resistance RF may vary with time. As a result, the fault current is
variable. Another common type of fault evolution is a change of fault type. Many faults
initiate as line-to-ground faults and evolve into line-to-line-to-ground faults and/or three-
phase-to-ground faults.
For evolving faults, it is generally difficult to detect fault inception and fault type
changes.
P, Q
21
V
V V2
Z = = Ðj = Ðj
I I S
Both phase and ground distance elements measure impedance for normal load conditions.
As can be seen in the figure, the measured impedance depends on the load flow
conditions. An increase in the apparent power, S, transferred over the line reduces the
magnitude of the measured impedance
X Maximum
Load
The impedance angle depends on the direction of P and Q. For positive values of P and Q
(both P and Q flowing into the line), j is between 0° and 90° and the measured
impedance lies on the first quadrant of the impedance plane. A negative Q value brings
the impedance to the fourth quadrant.
Accordingly, a negative P value will move the measured impedance to the other two
quadrants: to the second quadrant for a positive Q value and to the third quadrant for a
negative Q value.
The figure shows the possible regions of the measured impedance for normal load
conditions. For positive P values (Load OUT in the figure), the impedance is in the first
or fourth quadrants. For negative P (Load IN), the impedance is in the second or third
quadrants.
Load Encroachment
X
Checks for load encroachment problems can be performed by superimposing relay and
load characteristics on the impedance plane. In the figure, it can be seen that certain load
conditions result in penetration of the impedance in the relay characteristic.
A new solution is to create a relay load element having the same shape as the load
impedance regions. This new element may be used to block the distance element. In this
case, only a small section of the relay characteristic is lost.
Power Swings
Example Fault on Line A-B
A B
L1
q
The relay is supposed to be protecting line L1, and its location is at Bus P. The fault
occurs on line A-B.
Power Swings
No Relay Misoperation
Second Clearing
t=t2
X t=0+
First Clearing
t=t1 Fault
Pre-Fault
L1 t=0-
t=t3
p L1 Relay at p R
Post-Fault
Power Swing
The figure shows the impedance measured by a distance relay for an external fault. As
expected, the apparent impedance moves to an external point. After the fault clears, the
apparent impedance does not return instantaneously to the load equilibrium point, but
there is a post-fault “slow” oscillation, or swing. The characteristics of the oscillation
depend on many parameters of the power system. The case shown in the figure shows
that the oscillation does not produce any relay misoperation.
Power Swings
Relay Could Misoperate
Second Clearing
t=t2
X
First Clearing
t=0+
t=t1 Fault
Pre-Fault
L1 t=0-
t=t3
p L1 Relay at p R
Post-Fault
Power Swing
The figure shows a case where the post-fault power swing enters into the relay first zone
operation characteristic. This will produce an undesired operation of the relay.
Parallel lines is a very common case in transmission systems. The magnetic field
produced by a faulted line influences the behavior of the voltages and currents of a
neighboring line. This influence is called mutual coupling between lines.
I0M Z0M
IR,
I0
Relay
The figure illustrates mutual coupling. For a ground fault on one of two double-circuit
lines, the zero-sequence current flowing at each line induces a voltage in the other line.
This effect modifies the voltage ground distance elements measure. If the mutually
coupled current flows in the same direction as the relay current, the measured voltage
will increase and the distance element will tend to underreach. On the other hand, if
the mutual current direction is opposite to that of the relay current, the relay element
will measure a lower voltage and tend to overreach.
Magnetic mutual coupling between multiple-circuit lines affects distance and directional
ground relays. Typical values of the zero-sequence mutual coupling impedance, Z0M, are
on the order of 50 percent to 70 percent of the zero-sequence impedance, Z0. Very low
coupling occurs in the positive-sequence and negative-sequence networks.
( )
V a = m Z1L I a + k 0 I res + m Z0 M I 0 M
Va I0 M
Z= = m Z IL + m Z0 M
I a + k 0 I res I a + k 0 I res
These expressions show the effect of mutual coupling on the impedance ground distance
elements measure. I0M is assumed to be positive when it flows in the same direction as
Ires. The result is an increase in the apparent impedance, Z, and a relay underreach. A
negative value of I0M reduces the measured impedance, and the relay element
overreaches.
A B
Z
2 ZL ZC
ZL ZA
ZB
The figure shows the ground distance element behavior for a ground fault on a double-
circuit transmission line with both lines connected in parallel. For simplicity, it is
assumed that the system is energized at one end only.
In the lower figure, the impedances ground distance elements measure at locations A, B,
and C are shown as a function of the distance, m, to the fault from location A. For relay
elements A and C, the apparent impedances, with mutual coupling, are represented by
full lines, and dotted lines are used to represent the measured impedances without mutual
coupling. The latter case gives the correct impedance values and serves as a reference for
analysis. For relay element B, the measured impedance plot without mutual coupling
would be a straight line from ZB = ZL at m = 0, to ZB = 0 at m = 1.
The effect of mutual coupling in this case is that none of the ground distance elements
correctly measure the distance to the fault. For relay elements A and C, the mutually
coupled current flows in the same direction as the relay current. These elements measure
higher impedance values and tend to underreach. For relay element B, the mutual current
flows opposite to relay current. The result is a lower apparent impedance value and a
tendency for the relay element to underreach.
Zero-Sequence
Mutual Coupling Compensation
( )
Va = mZ1L Ia + k0 Ires + mZ0M I0M
æ Z0 M ö
ç
V a = m Z1L I a + k 0 I res + I0 M ÷
ç Z1L ÷
è ø
I relay = I a + k 0 I res + k 0 M I0 M
k 0 M = Z0 M Z1L
The mutual-coupling error may be compensated for by providing the faulted-line relay
element with information on the mutually coupled current. These equations show that an
additional compensation term, k0M I0M, is needed in the relay current of the ground
distance element. The compensation factor, k0M, equals the ratio of the zero-sequence
impedance to the positive-sequence impedance of the protected line. Several commercial
distance relays have an additional current input for the mutually coupled current.
Mutual Coupling
Compensation Limitations
In the parallel-line case that was presented before, compensation works for relays at
locations A and B and fails for the relay at C. The problem with the relay at C is that the
level of compensation needed depends on the fault location, m.
Another limitation of mutual coupling compensation is that I0M is not always available at
the relay location.
This figure shows cases in which the mutually-coupled current I0M is not available at
the relay location.
Experience shows that the first zone setting for the relay should be reduced to avoid
overreaching for severe cases of zero-sequence mutual coupling. In some cases, the first
zone may need to be reduced by 60 percent.
Introduction to
Distance Relay Polarizing
Signal S1
V Forming Phase
Trip
I Comparison
Signal S2
Forming
S1 = -V + Zr I Operating Quantity
S2 = V Polarizing Quantity
Mho Characteristic
Zr Zr-Z
90o
A R
X F4
S1 = - V + Z r I
F3 S2 = V
S1 Phase Changes
F1 S2 Phase Does Not Change
F2 R
Loss-of-Polarization
for Close-In Bolted Faults
X
V I
F2 F1
21 F1
F2 R
V = 0, S2 = 0
Polarization Alternatives
l Self-Polarization
l Dual-Polarization
t Addition of healthy phase voltage
l Cross-Polarization
l Positive-Sequence
l Memory
t Three-phase faults, voltage reversals
Self-Polarization: S 2 = V
S1 = - V + Z r I S2 = V + k p V p
æ Z - Zr ö
0 £ argç ÷£p
çZ+Z ÷
è p ø
A = -Z p = - k p V p I B = Zr
1 2 F 3
X 3
B
Zr Z r
2
A R
A'
1
1 F 2 3
3
X B
A''
2 Zr
A R
1
l Self-Polarization
l Healthy-Phase Cross
Polarization (With and Without
Memory)
l Positive-Sequence Polarization
(With Memory)
S1 = - V bc + Zr Ibc
S 2 = V bc
l No Characteristic Expansion
l Unreliable for Zero-Voltage Faults
l Directionally Insecure for Reverse Bus
Faults During High Load. Requires
Additional Directional Element
(
S1 = - Va + Zr Ia + k 0 I0 )
S2 = Va
l No Characteristic Expansion
l Unreliable for Zero-Voltage Single
Line-to-Ground Faults
l Requires Directional Element
S1 = - V bc + Zr Ibc
S2 = - j Va
l Good Expansion for Phase-to-Phase
Faults
l Unreliable for Zero-Voltage Three-Phase
Faults
l Requires Directional Element
(
S1 = - Va + Zr I a + k 0 I 0 )
S 2 = j Vbc
l Good Expansion
l Reliable Operation for Zero-Voltage
Single Line-Ground Faults
l Requires Directional Element
l Single-Pole Trip Applications Require
Study for Pole-Open Security
Phase Element
Cross Polarization With Memory
S1 = - Vbc + Zr Ibc
S2 = -j Va,mem
l Good Expansion
l Reliable Operation for Zero-Voltage
Three-Phase Faults
l Requires Directional Element
l Single-Pole Trip Applications
Require Study for Pole-Open
Security
l Greatest Expansion
l Reliable Operation for Zero-Voltage
Three-Phase Faults
l Requires Directional Element
l Best Single-Pole Trip Security
(
S1 = - Va + Zr Ia + k 0 I0 )
S2 = Va1,mem
l Greatest Expansion
l Reliable Operation for Zero-Voltage
Faults
l Requires Directional Element
l Best Single-Pole Trip Security